# Every Step Counts: Decoding Trajectories as Authorship Fingerprints of dLLMs

arXiv: https://arxiv.org/abs/2510.05148 · HTML: https://mlsanity.com/paper/2510.05148/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Discrete and text diffusion, one of 14 directions in Diffusion models.
- Builds on 24 of its 43 references that are in the map.
- First posted on arXiv 2 Oct 2025.

## Placed in

- Diffusion models / [Discrete and text diffusion](https://mlsanity.com/diffusion/discrete-and-text-diffusion/)

## Builds on (most cited first, 24 of 24)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature](https://mlsanity.com/paper/2301.11305/) (2023)
- [A Watermark for Large Language Models](https://mlsanity.com/paper/2301.10226/) (2023)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [Detecting Language Model Attacks with Perplexity](https://mlsanity.com/paper/2308.14132/) (2023)
- [All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text](https://mlsanity.com/paper/2107.00061/) (2021)
- [Automatic Detection of Machine Generated Text: A Critical Survey](https://mlsanity.com/paper/2011.01314/) (2020)
- [TweepFake: about Detecting Deepfake Tweets](https://mlsanity.com/paper/2008.00036/) (2020)
- [Perplexed by Perplexity: Perplexity-Based Data Pruning With Small Reference Models](https://mlsanity.com/paper/2405.20541/) (2024)
- [Detecting Multimedia Generated by Large AI Models: A Survey](https://mlsanity.com/paper/2402.00045/) (2024)
- [Discrete Diffusion in Large Language and Multimodal Models: A Survey](https://mlsanity.com/paper/2506.13759/) (2025)
- [A Systematic Review on Model Watermarking for Neural Networks](https://mlsanity.com/paper/2009.12153/) (2020)
- [Spear Phishing With Large Language Models](https://mlsanity.com/paper/2305.06972/) (2023)
- [Origin Tracing and Detecting of LLMs](https://mlsanity.com/paper/2304.14072/) (2023)
- [The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMs](https://mlsanity.com/paper/2507.11097/) (2025)
- [Your Large Language Models Are Leaving Fingerprints](https://mlsanity.com/paper/2405.14057/) (2024)
- [From Text to Source: Results in Detecting Large Language Model-Generated Content](https://mlsanity.com/paper/2309.13322/) (2023)
- [Jailbreaking Large Language Diffusion Models: Revealing Hidden Safety Flaws in Diffusion-Based Text Generation](https://mlsanity.com/paper/2507.19227/) (2025)
- [Detecting Machine-Generated Texts: Not Just "AI vs Humans" and Explainability is Complicated](https://mlsanity.com/paper/2406.18259/) (2024)
- [Where to Start Alignment? Diffusion Large Language Model May Demand a Distinct Position](https://mlsanity.com/paper/2508.12398/) (2025)
- [An Investigation of Supervised Learning Methods for Authorship Attribution in Short Hinglish Texts using Char & Word N-grams](https://mlsanity.com/paper/1812.10281/) (2018)
- [Who could be behind QAnon? Authorship attribution with supervised machine-learning](https://mlsanity.com/paper/2303.02078/) (2023)

## Built on by (most cited first, 4 of 4)

- [Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models](https://mlsanity.com/paper/2602.22246/) (2026)
- [Anatomy of a Lie: A Multi-Stage Diagnostic Framework for Tracing Hallucinations in Vision-Language Models](https://mlsanity.com/paper/2603.15557/) (2026)
- [AutoMIA: Improved Baselines for Membership Inference Attack via Agentic Self-Exploration](https://mlsanity.com/paper/2604.01014/) (2026)
- [Backdooring Masked Diffusion Language Models](https://mlsanity.com/paper/2605.19262/) (2026)

## Abstract (arXiv)

Discrete Diffusion Large Language Models (dLLMs) have recently emerged as a competitive paradigm for non-autoregressive language modeling. Their distinctive decoding mechanism enables faster inference speed and strong performance in code generation and mathematical tasks. In this work, we show that the decoding mechanism of dLLMs not only enhances model utility but also can be used as a powerful tool for model attribution. A key challenge in this problem lies in the diversity of attribution scenarios, including distinguishing between different models as well as between different checkpoints or backups of the same model. To ensure broad applicability, we identify two fundamental problems: what information to extract from the decoding trajectory, and how to utilize it effectively. We first observe that relying directly on per-step model confidence yields poor performance. This is mainly due to the bidirectional decoding nature of dLLMs: each newly decoded token influences the confidence of other decoded tokens, making model confidence highly redundant and washing out structural signal regarding decoding order or dependencies. To overcome this, we propose a novel information extraction scheme called the Directed Decoding Map (DDM), which captures structural relationships between decoding steps and better reveals model-specific behaviors. Furthermore, to make full use of the extracted structural information during attribution, we propose Gaussian-Trajectory Attribution (GTA), where we fit a cell-wise Gaussian distribution at each decoding position for each target model, and define the likelihood of a trajectory as the attribution score: if a trajectory exhibits higher log-likelihood under the distribution of a specific model, it is more likely to have been generated by that model. Extensive experiments under different settings validate the utility of our methods.